MarketPsych Insights | Book Review

Behavioral Investment Management: An Efficient Alternative to Modern Portfolio Theory

by Greg B. Davies and Arnaud De Servigny (McGraw-Hill, 2012)

"People in standard finance are rational. People in behavioral finance are normal."

-Meir Statman, PhD, Santa Clara University

The above quote by Meir Statman captures concisely the thrust of the concepts explored extensively in this new book by Greg Davies and Arnaud De Servigny. They tackle what is by now a relatively widespread view that modern portfolio theory (MPI) has fundamental flaws when used blindly to build portfolios for individuals. However, rather than throwing the baby out with the bath water, they zero in with quantitative precision on the aspects of this developed theory that they counsel should be modified to reflect the fundamental truth of how humans process concepts and make decisions.

You should note that Greg was, at the time this book was published, Head of both Behavioral and Quantitative Investment Philosophy at Barclays Wealth. Arnaud was the Global Head of Discretionary Portfolio Management and Investment Strategy at Deutsche Bank Private Wealth Management and an Adjunct Professor of Finance at Imperial College Business School in London. Given their backgrounds, both behavioral and statistical topics are explored in this book with extensive use of quantitative methods (think lots of formulas).

The first four chapters review the concepts of modern portfolio theory (MPT), efficient market hypothesis (EMH), and expected utility theory (EUT). Additionally, they provide their take in detail on the flaws of using them for individual investor portfolio optimization. They indicate that defining an optimum portfolio for an individual requires that the investor -

  • Have a clear notion of their preferences for risk and return; and
  • Make assessments of expected risk and return of the investment assets reasonably accurately.

We may all agree, but finding clarity and agreement with clients on these two points is inordinately challenging.

The authors acknowledge that individuals need to make two decisions. First, what portion of wealth should be consumed in the immediate future versus saved to fund longer-term needs. Second, how the savings should be optimally invested. The book focuses on the second question, but (in my opinion) working through the process to identify answers to the first can be remarkably helpful in diagnosing some of the biases and normal emotional pitfalls that are addressed later in the book.

A few other insightful observations they suggest to "behavioralize" finance are:

  • Time horizons are interpreted by people as emotional time horizons, not practical ones.
  • Many people focus inordinately on a single investment decision versus its impact on the overall portfolio.
  • Citing cumulative prospect theory[1], they explain that people focus inordinately on changes in wealth levels from current values versus their overall level of wealth (which should not be used to guide investment decisions).
  • The explanation of the need to incorporate two mental systems (sometimes referred to as the preference for deliberative versus intuitive decision making process, or system 1 and system 2 thinking), they explain simply as the long term "rational" self and it's short term irrational brother.

A few practical suggestions are provided aimed at offsetting these biases to mitigate reactivity:

  • Express figures in annual or five year terms – in order to lengthen the time horizon for decision framing.
  • Reduce the frequency of monitoring performance.
  • Engage a process where investors pre-commit to a planned investment decision that will be executed automatically.

Chapter Five (which included contributions by Shweta Agarwal, a PhD student at the London School of Economics) contains several key concepts in the formulation of amendments to traditional asset allocation work. Specifically, the more tailored inclusion of risk in portfolio construction. The key change is to adjust the typical asset categorical risk factor (i.e., variance) to include an assessment of how much return (in excess of the risk free rate) it would take to justify an investment. This adaptive measure is termed behavioral standard deviation. The examples illustrate the portfolio construction decisions for nine asset categories of three sample individuals based on their personalized behavioral standard deviation (high risk tolerance, moderate or "normal", and high anxiety/low risk tolerance).

The implication is that in a world of investment possibilities that include a myriad of non-normal performance possibilities (think skewness and kurtosis), we can and should identify an efficient frontier to maximize returns for an individual's specific level of behavioral risk.

Chapters six and seven provide insights on how to view the process of evaluating market dynamics and portfolio construction through the lens of behavioral implications. One interesting observation is that their data support the concept of monthly rebalancing of portfolios to align with strategic policy allocations. Additionally, the concept of more active and frequent assessment of expected returns and reshaping the optimized portfolio are inherent in their approach.

Chapter eight tackles what are perhaps two most impactful set of issues in dealing with individual investors; namely, typical errors or biases, and typically an inordinate dependence on unstable short-term preferences. The authors propose that both be removed from the process of portfolio optimization. While you may or may not agree with this conclusion, my experience is that addressing them is highly challenging. My sense is that an emerging number of researchers believe that biases simply cannot be eliminated in individuals, but only recognized and marginalized when it is clear they adversely impact the efficiency and effectiveness of the decision-making process.

Strategies that they propose for addressing these issues are:

  • Align decision frames with long-term objectives (to enhance the impact of a more realistic/longer time horizon).
  • "Purchase emotional comfort". The concept is that we might develop a range of prescriptive actions as salve to alleviate our burning emotions. The idea includes identifying costs relative to objectives. (i.e., maintaining a large cash reserve for peace of mind is balanced against the foregone returns in deciding how much "medicine" to take).
  • Increase allocations to low-anxiety investments and engage downside protection strategies. While some of these strategies may come with costs, they may provide sufficient benefit to remove undue emotional influence on decisions. Specifically consider:
    • Option strategies to eliminate far-left tail exposure.
    • Funds that have a longer time horizon or are less frequently marked-to-market for valuations.
    • Investments with narrower downside exposure to monthly losses.
    • Funds that bundle a wide range of investments in their reporting (aggregation = less volatility).
    • Dollar cost averaging for investments (of value when a large investment is being made).
    • Avoiding information (or media) too focused on short-term or detailed asset price movements.
    • Engaging a sign off process for big investment decisions that automatically incorporate deliberative thinking (perhaps by independent advisors). Alternatively, require a cooling off period prior to executing a decision.

The final chapter in their book includes ideas that from their perspective would help enhance the state of the wealth management industry by incorporating their ideas in tailoring the process more adroitly to client needs. Specifics are:

  • Advisors are (typically) risk managers for their clients. This role requires more substantive attention to focus on appropriate tools to profile and diagnose client's individual risk preferences and incorporate them into their consulting support.
  • In this role, risks to be addressed are structural (markets/investment structures), emotional, and liquidity. Specifically, in a world in which our demographics are shifting to enhanced concerns about managing living expenses over remaining lifetimes, this is an area of growing importance.
  • Move from static asset allocation methods to those that are more dynamic. They note that the evaluation horizon for wealthy people is moving toward a 2 to 3-year horizon versus 5 to 10 as in the past.
  • They foresee firms which adapt a more behaviorally focused approach to client service processes will ultimately achieve a significant competitive advantage.

While this book is not an easy read, it is packed with practical insights of great value to advisors who wish to undertake a more substantive incorporation of behavioral techniques in their client serving process.

  1. Kahneman and Tversky (1979), Tversky and Kahneman (1992) ↩︎